shuttleai/shuttle-2.5-mini

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:12BQuant:FP8Ctx Length:32kPublished:Jul 27, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Warm

Shuttle-2.5-mini is a 13 billion parameter multilingual language model developed by ShuttleAI Inc., fine-tuned from Mistral-Nemo-Base-2407. It is specifically designed to excel in complex chat, reasoning, and agent tasks, with a unique focus on emulating the writing style of Claude 3 models and extensive training on role-playing data. This model supports a 128k context window and is pretrained on a large proportion of multilingual and code data, making it suitable for diverse communication and specialized interactive applications.

Loading preview...

Shuttle-2.5-mini Overview

Shuttle-2.5-mini is a 13 billion parameter multilingual language model developed by ShuttleAI Inc., released under the Apache 2.0 License. It is a fine-tuned version of the Mistral-Nemo-Base-2407 model, specifically engineered to excel in complex chat, reasoning, and agent tasks. A key differentiator is its extensive training to emulate the writing style of Claude 3 models and its thorough fine-tuning on role-playing data, making it highly adept at nuanced and interactive conversational scenarios.

Key Capabilities

  • Claude 3 Style Emulation: Fine-tuned to mimic the prose quality and conversational style of Claude 3 models.
  • Role-Playing Proficiency: Extensively trained on role-playing data for highly engaging and context-aware interactions.
  • Multilingual Support: Pretrained on a significant proportion of multilingual data, supporting diverse language applications.
  • Extended Context Window: Features a 128k context window, allowing for processing longer and more complex inputs.
  • Reasoning and Agent Tasks: Designed to perform well in complex reasoning and agent-based applications.

Good for

  • Advanced Chatbots: Ideal for creating chatbots that require sophisticated conversational styles and role-playing abilities.
  • Multilingual Applications: Suitable for applications needing robust performance across various languages.
  • Interactive Storytelling & Gaming: Excels in scenarios requiring dynamic character interactions and narrative generation.
  • Agent-Based Systems: Useful for developing AI agents that need to understand and respond within specific contexts and personas.

Popular Sampler Settings

Top 3 parameter combinations used by Featherless users for this model. Click a tab to see each config.

temperature
top_p
top_k
frequency_penalty
presence_penalty
repetition_penalty
min_p